Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Nov 23:6:37393.
doi: 10.1038/srep37393.

Changes of bivalent chromatin coincide with increased expression of developmental genes in cancer

Affiliations

Changes of bivalent chromatin coincide with increased expression of developmental genes in cancer

Stephan H Bernhart et al. Sci Rep. .

Abstract

Bivalent (poised or paused) chromatin comprises activating and repressing histone modifications at the same location. This combination of epigenetic marks at promoter or enhancer regions keeps genes expressed at low levels but poised for rapid activation. Typically, DNA at bivalent promoters is only lowly methylated in normal cells, but frequently shows elevated methylation levels in cancer samples. Here, we developed a universal classifier built from chromatin data that can identify cancer samples solely from hypermethylation of bivalent chromatin. Tested on over 7,000 DNA methylation data sets from several cancer types, it reaches an AUC of 0.92. Although higher levels of DNA methylation are often associated with transcriptional silencing, counter-intuitive positive statistical dependencies between DNA methylation and expression levels have been recently reported for two cancer types. Here, we re-analyze combined expression and DNA methylation data sets, comprising over 5,000 samples, and demonstrate that the conjunction of hypermethylation of bivalent chromatin and up-regulation of the corresponding genes is a general phenomenon in cancer. This up-regulation affects many developmental genes and transcription factors, including dozens of homeobox genes and other genes implicated in cancer. Thus, we reason that the disturbance of bivalent chromatin may be intimately linked to tumorigenesis.

PubMed Disclaimer

Figures

Figure 1
Figure 1. Bivalent chromatin segments in normal and cancer tissues.
(A) Cancer cell lines show less bivalent regions than other types of cells. Number of bivalent chromatin segments in cancer cell lines (red), normal tissues or cells (grey), embryonic stem cells (ESC, blue) and induced pluripotent stem cells (iPSC, light blue). Stem cells on average have the highest number of bivalent segments, while cancer cell lines have low numbers of bivalent segments. (B) GO-term enrichment for genes associated to bivalent chromatin segments. Light grey: genes associated to any bivalent segment; Dark grey: Genes associated to a frequently bivalent segment (FBS genes). The enrichment of metabolic process genes is lost in FBSs, while the FBS genes are stronger enriched in the GO-terms developmental processes, anatomical structure morphogenesis, (*) sequence-specific DNA binding RNA polymerase II transcription factor activity and signaling. (C) FBSs well reflect normal bivalent chromatin states. Left: Fraction of bivalent chromatin segments that are part of a FBS in cancer cell lines (red), normal tissues or cells (grey), ESC (blue) and iPSC (light blue). On average, cancer cells are depleted in bivalent chromatin overlapping FBSs. Right: Fraction of FBSs that are bivalent in cancer cell lines (red), normal tissues or cells (grey), ESC (blue) and iPSC (light blue). FBSs are infrequently bivalent in cancer cell lines compared to normal cells. (D) Length distribution of FBSs (black) and segments based on active promoters (red) and enhancers (blue).
Figure 2
Figure 2. Cancer cell lines lose bivalent chromatin.
Comparison of stability of chromatin state segments between related normal cells (grey) and cancer cell lines and their cells of origin (red). Stability described as relative number of nucleotides of the respective chromatin type where type is the same in both cells. Bivalent segments (Biv) are the least stable chromatin state segments in cancer cell lines, while they show average stability in related normal cells. Biv: Bivalent chromatin, Enh: Enhancer, Hetchr: Heterochromatin, Quies: Quiescent, TssA: Active promoter, Tx: Transcribed, ZNF: Zinc-finger/insulator, ReprPC: Polycomb repressed.
Figure 3
Figure 3. Hypermethylation of bivalent chromatin in cancer tissues.
(A) Loss of bivalency is associated to hypermethylation. DNA methylation change of bivalent segments in relation to their stability. DNA methylation change for fresh cancer (left) and cancer cell lines (right) is shown for all elements in the cells of origin. Violet: chromatin bivalent in cancer and cell of origin, blue: chromatin no longer bivalent in cancer, grey: chromatin bivalent in cancer but not in cell of origin. (B) Bivalent chromatin regions show highest hypermethylation. DNA methylation difference for the 15 Roadmap chromatin states in BL. Bivalent states (grey box) show the strongest hypermethylation. (C) FBS regions are hypermethylated along their whole length. DNA methylation difference in BL at (grey marked area) and around FBSs, shown for FBSs > 10,000 nt (black), FBSs < 10,000 nt (blue), FBS sub-segments bivalent in primary B cells (red) and segments bivalent in primary B cells but not overlapping FBS (brown). (D) Methylation of bivalent regions is a good descriptor of cancer. ROC curve using relative descriptor methylation of FBSs (orange), bivalent enhancer (black), bivalent promoters (green), polycomb repressed regions without FBS (ocher) and ESC derived classifier (blue).
Figure 4
Figure 4. Up-regulation of bivalent chromatin genes in cancer tissues.
(A) Scatter plots of DNA methylation change versus expression change for significantly differentially expressed FBS genes (left) and controls: active promoter background (center) and chromatin bivalent in ESC exclusively (right). (B) Cumulative plot of the number of cancers and gene/promoter pairs in quadrant one (Q1, hypermethylation and up-regulation) as the fraction of FBS genes (blue), ESC genes (grey) and background (red). Top: Only significantly differentially expressed genes are considered. Bottom: Only lowly and differentially expressed genes are considered. (C) Box-plot showing the expression change of FBS genes (blue), genes bivalent in ESC (grey) and other genes (red). Significance (Wilcoxon) of comparisons: ***p < 2.2 × 10−9, **p < 10−3, *p < 0.05. BL: Burkitt’s Lymphoma, BLCA: Bladder Urothelial Cancer, BRCA: Breast Cancer, CESC: Cervical Squamous Cell Carcinoma, COAD: Colon Adenocarcinoma, ESCA: Esophagus Cancer, FL: Follicular Lymphoma, HNSC: Head and Neck Squamous Cell Carcinoma, KIRC: Kidney Renal Clear Cell Carcinoma, KIRP: Kidney Renal Papillary Cell Carcinoma, LIHC: Liver Hepatocellular Carcinoma, LUAD: Lung Adenocarcinoma, LUSC: Lung Squamous Cell Carcinoma, PAAD: Pancreas Adenocarcinoma, PRAD: Prostate Adenocarcinoma, READ: Rectum Adenocarcinoma, SKCM: Skin Cutaneous Melanoma, THCA: Head and Neck Thyroid Carcinoma, UCEC: Uterine Corpus Endometrial Carcinoma.
Figure 5
Figure 5. Expression of FBS genes.
(A) Combined scatter and boxplots of expression differences in qPCR experiments for DLX5, LEF1, PAX2, SP8 and ZIC2 genes for 20 HCC cancer samples against normal liver tissue. The red dots show the log of the mean fold change, which is positive for all cancer experiments. (B) Heatmap of the expression log2 fold changes of selected FBS genes. Unsupervised clustering reveals expression similarities of FBS genes in related tissues. BL: Burkitt’s Lymphoma, BLCA: Bladder Urothelial Cancer, BRCA: Breast Cancer, CESC: Cervical Squamous Cell Carcinoma, COAD: Colon Adenocarcinoma, ESCA: Esophagus Cancer, FL: Follicular Lymphoma, HNSC: Head and Neck Squamous Cell Carcinoma, KIRC: Kidney Renal Clear Cell Carcinoma, KIRP: Kidney Renal Papillary Cell Carcinoma, LIHC: Liver Hepatocellular Carcinoma, LUAD: Lung Adenocarcinoma, LUSC: Lung Squamous Cell Carcinoma, PAAD: Pancreas Adenocarcinoma, PRAD: Prostate Adenocarcinoma, READ: Rectum Adenocarcinoma, SKCM: Skin Cutaneous Melanoma, THCA: Head and Neck Thyroid Carcinoma, UCEC: Uterine Corpus Endometrial Carcinoma. (C) Heatmap of expression changes for all FBS genes significantly up-regulated in more than 50% of the cancer data sets.
Figure 6
Figure 6. Expression and DNA methylation of cancer cell lines.
(A) DNA methylation and expression changes of FBSs in BL cell lines against BL fresh tissues (top left), KARPAS-422 against FL fresh tissues (top right), lymphoma cell lines against normal germinal center B-cells (bottom left) and germinal center B-cells against FL fresh tissues. Changes are shown relative to the cell lines. The number of pairs in each quadrant is shown. (B) Expression change of FBS genes (blue) and other genes against normal cells in glioblastoma fresh tissue (left) and in cancer cell lines after a number of passages. At the right, the value for commonly used cell lines is shown. (C) Relative methylation of classifier FBSs in cancer cell lines, fresh cancer tissue, normal cell lines and normal tissue.

References

    1. International Cancer Genome Consortium. International network of cancer genome projects. Nature 464, 993–998 (2010). - PMC - PubMed
    1. Cancer Genome Atlas Research Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 455, 1061–1068 (2008). - PMC - PubMed
    1. Roadmap Epigenomics Consortium et al.. Integrative analysis of 111 reference human epigenomes. Nature 518, 317–330, doi: 10.1038/nature14248 (2015). - DOI - PMC - PubMed
    1. Mikkelsen T. S. et al.. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448, 553–560 (2007). - PMC - PubMed
    1. Voigt P., Tee W.-W. & Reinberg D. A double take on bivalent promoters. Genes & development 27, 1318–1338 (2013). - PMC - PubMed

Publication types

MeSH terms